The use of location-based services in maintaining electrical and other networks in developing countries
Barry Dwolatzky Rex van Olst, Modiri Seate University of the Witwatersrand PO Box 452, Wits, 2050 South Africa Abstract: In South Africa and other developing countries there is a major drive to expand infrastructure. Over the past 10 years millions of South African households in both rural and urban areas have been provided with electricity, water and telephone services. Once new networks have been installed the challenge is to maintain them . Maintenance personnel typically have a lower level of experience and skill than their counterparts in Western Europe and the USA. This paper describes research into the application of Location-based Services [LBS] technology to support work teams responsible for maintaining network-based utilities in developing countries. By providing work teams with mobile hand-held computing devices linked via a communication network to a central geospatially-based data server, information about the network can be provided as required. In addition the proposed system compensates for the work team's lack of familiarity with computers by providing a context aware human-computer interface. This limits the amount of data input required from the operator. The focus of the paper is on the maintenance of electrical distribution networks, but the principles explored are applicable to telecommunications and water networks in under-developed areas. Intorduction In 2002 most African countries came together to form a new political and economic bloc called the “African Union”, or AU. The economic and developmental aspirations of the AU are encapsulated in a strategic plan called “NEPAD” (New Economic Plan for Africa’s Development). One of the key aspects of NEPAD is infrastructural development. Bridging the infrastructure gap in Africa has been identified as an important element of promoting regional integration on the continent [NEPAD, 2002]. This strategy envisages massive investments in the construction of new electrical, water and telecommunication networks. Infrastructural growth has also been a key feature of the development of South Africa since the early 1990’s. More than 2 million households in both urban and rural areas have been supplied with grid electricity and piped water over the past 10 years. Fixed-line and cellular telecommunications networks are being rapidly rolled-out. The impressive rate of growth of network-based utilities in South Africa has, however, brought with it many challenges. It is these same challenges that many African and other developing countries are bound to face in the years ahead. This paper deals with one of these challenges – namely the task of operating and maintaining hundreds of thousands of kilometers of new network infrastructure. Network Maintenance - The Challenge Maintenance personnel in developing countries typically have a lower level of experience and skill than their counterparts in Western Europe and the USA. In bridging this skills gap, modern information and communication technologies has the potential to make a significant contribution. One such technology relates to spatial information systems (SIS). In the developed world, as SIS technology matures, it is increasingly being applied to a variety of new tasks within utilities, including planned maintenance, the real-time location of faults on the network, and the dispatch of maintenance vehicles to rectify those faults. In relying on SIS technology to perform these tasks, managers are discovering an ever-widening range of other applications for which the technology is suitable. In developing countries SIS technology has the potential to empower the mobile workforce. The appropriate use of computer applications based on geospatial information will improve the ability of inexperienced and relatively unskilled field personnel, allowing them to reduce costs, improve productivity and accuracy, and be more responsive to customers and clients [Johnson, 1998]. According to Scott Johnson, there are three primary ways of improving productivity of a workforce. These are i) through the development of new technologies; ii) through increased capital expenditure; and iii) through education and training. All of these are pertinent in Southern Africa and to the theme of this paper. However although the development of technology is a key to productivity improvement, the technology is worthless unless it is actually used. Whilst great improvements in productivity will be unlikely unless workers have the level of education and skill needed to handle the new (advanced) technologies. The theme of this paper is to ensure “ease-of-use” of the technologies proposed. The application of SIS technology, whether in the developed or developing world, is however pointing the way to a need for enterprise wide access to spatial data embedded in the corporate database. The implication of the demand for user access to spatially related data (asset maps, geocoded networks, field workprints and orders, dispatch, and topographical maps etc) requires that the corporate relational data model includes two simple additional attributes, namely "x" and "y" map co-ordinates, for any outside plant, personnel and customers [Lancaster, 1996]. Almost all utilities have to face the challenge of capturing and maintaining field asset records, which are required in an "as-built" format to supply accurate information to work teams in the field. With geospatial information content now more easily available it is possible to capture, maintain and access field asset data with the assistance of personal digital assistants (PDAs) and wireless communications technology, thus enhancing the productivity of field operational work-crews. The dramatic growth in business demand for mobile computing is a direct result of the natural synergy between PDAs, wireless technology, and data-driven workstyles [Intel, 2002]. Multiple studies of this synergy confirm productivity gains of between 15 and 25% per week when the mobile workforce is equipped with mobile PDAs and wireless access. The Gartner Group, in one of its studies, found that the amount of time wasted by the average worker on paper-related tasks, regardless of their industry, was 30% of each day ! To deal with the key challenge of network data collection our University is carrying out research focused on developing appropriate business processes to achieve more productive data capture. This research involves the measurement of existing data capturing productivity within South Africa’s telecommunications company, Telkom, and the national power utility, Eskom. Best practice, worst practice and something in between are to be measured as a "current status" baseline. The "quantifiable increase in productivity" will be measured by applying the proposed business process and technology to a pilot site and then assessing productivity gains over the baseline. In a similar way data accuracy improvements will be quantified. These measurements will be conducted in an “over-the-shoulder” fashion with Telkom’s and Eskom’s own personnel to minimise duplication and to ensure that the measurements are not “skewed” by introducing alien processes through the utilisation of only non-Telkom or Eskom field personnel for the data capture implementation. Standard COTS (commercial-off-the-shelf) software will be used in the new data capturing process proposed, and where appropriate, the field personnel will be trained in its use. This software is Intergraph’s package called Intelliwhere OnDemand, software that operates on a standard Personal Digital Assistant (PDA) device to location-enable a mobile workforce. This package will be customised to accommodate the specific requirements of the data capturing application envisaged. It is planned to deploy the COMPAQ iPAC Pocket PC PDA as the data capture device itself. Human Computer Information(HCI) By suggesting the use of SIS as a way of compensating for the lack of skill and experience amongst field personnel in developing countries, we bring another significant challenge to the fore. This is the low level of familiarity with computers amongst field personnel. In this regard our research is addressing three additional questions:
The development of mobile computers able to communicate via wireless technology has brought a completely different perspective and approach on how humans communicate with computers. Interaction with mobile computers is vastly different as compared to desktop computers. Desktop computing takes place in a static environment where the user is presented with a large high-resolution user interface and information is stored either locally on the computer, or may be accessed via a data network. [Cobb, 2002] states that “usability” is important for any mobile computing application or device, but it’s even more important with capability-constrained devices with limited display or input. Thus usability analysis should encompass any wireless/PDA application project from start to completion. In contrast, mobile computing has associated with it specific constraints and restrictions which result in the design of very different user interfaces. Two of the most important limitations of mobile computing are:
The topic of “adaptive intelligent user interfaces” in the realm of Human Computer Interaction [HCI] has been researched in recent years in both mobile and static computing environments. In mobile computing, the focus has been on the services that adaptive mobile systems should provide and on adaptive frameworks to assist mobile navigation systems in adapting to limited resources [Baus, 2001]. Strategies are proposed on how to maintain effective HCI interaction when various resources become too limited to allow normal interaction to continue [Cheverst, 2000]. Environment sensing technologies have been used to develop adaptive user interfaces for ultra-mobile devices such as PDAs, mobile phones and wearable computers [Schmidt, 1999]. The objective of much of this work has been to illustrate the application of context-awareness in ultra-mobile computing. An example is an orientation-sensitive user interface, where a PDA is enhanced with awareness of its orientation. The field of adaptive user interfaces is vast, and there are many areas of potential research still to be exploited. Field Maintenance Information and workflow In our research we have developed an approach to the presentation of information to the user of a mobile device, with minimal user input. Our approach is based on an initial analysis of the information required by a work team member engaged in the maintenance of electrical and other networks. Electrical maintenance teams perform diverse tasks, including sectionalising, re-routing, fault-finding, planned maintenance, etc. The main objectives are to ensure that customers have constant power supply and that the network operates in a normal and stable mode. A centralised distribution control centre manages all network-related duties by issuing operating instructions to the teams whilst on site. The Work Management Centre (WMC) dispatches the teams to various areas and issues them with the details of the customers with no power supply. Circumstances occur when the network experiences disturbances, leading to the protection systems isolating the affected portion of the network e.g. overloading due to a lightning strike. This is termed a breakdown. Network stabilisation and fault rectification have to be performed within a particular timeframe in order to minimise damage to the network components (transformers, cables etc) and reconnect the affected customers. In these cases Control Centre operators send the Work Management Center a list of the affected network points, e.g. the locations of all open circuit breakers. The WMC dispatches this information to the teams in the affected areas. Planned maintenance of network components is also performed on pre-arranged schedules. On a specific day, based on these schedules, the Control Centre issues operation instructions to the work team to isolate the specific portion of the network for maintenance purposes. The work team’s tasks are not necessarily performed in a predetermined sequence. Based on a situation at hand, the team might temporarily postpone a task and perform another more urgent one. The team might also operate at a location on the network common to a number of disconnected customers, instead of attending to each customer individually. The basis for such a decision is detailed information on the network configuration. When a customer reports a fault, the Customer Service Centre captures specific customer details including the customer stand number, contact details and description of the problem e.g. no supply. The information is automatically sent to the Work Management Centre. At the WMC, a work-order number is generated for each reported fault. This number is used by the WMC to keep track of the team’s work progress on reported faults. In a breakdown scenario, the Control Centre employs a FMS (Fault Management System) to detect whether the circuit breakers have tripped or not. The Control Centre can electronically open and close the breakers. If a breaker trips after being electronically closed, the FMS automatically sends the circuit breaker information to the WMC. The information includes the name and location of the circuit breaker. The WMC dispatcher allocates a work-order number to a team in the same area as the customer. The team must initially accept the work order and then proceed to the network location. After completing a task, the team contacts the WMC and quotes the work-order number for the completed work at the network location. The dispatcher then updates the status of the particular work-order number, i.e. the maintenance job for the work-order number is complete or still in progress. Underlying concept: The concept of an adaptive context, information self-triggering HCI is employed to facilitate the development of the user interface for a mobile device. The concept is tailored from the established concepts of context awareness and triggering information by context, utilised in previous and on-going research work [Brown, 1998] [Rodden, 1998]. Context awareness in mobile computing deals with the ability of a mobile device to be aware of the context within which it is used and provide relevant information and/or services to the user, where relevancy depends on the user’s task. Various technologies have been used in context aware applications to acquire contextual information– e.g. sensors, active badges. Contextual information may include aspects such as time, location, speed and details about nearby objects. Adaptive Context User Interface: The concept of the “adaptive context user interface” puts emphasis on the ability of the user interface to adapt to the context within which the mobile device is used. Some researchers have argued that there is more to context than location in context-aware mobile computing [Schmidt, 1999]. The argument is that beyond location, awareness of further features that contribute to context, in particular the awareness of the physical conditions in a given environment enhances ultra-mobile devices to provide better support to their users. In the case of maintenance work teams, however, the context within which they utilise the mobile device is primarily location. Physical conditions such as temperature, time, orientation etc. do not contribute to services that the devices will provide. Since the work team frequently moves from one network point to another within their allocated area of operation, the application adapts to these movements by displaying details of the surrounding area relevant to the user’s current position in the area. Thus, the user interface displays the surrounding area map, including the network layout, relevant to the team’s position. The adaptation is achieved by incorporating GPS functionality with real time moving map technology. The customers in urban and rural areas are quite close to each other. A set of GPS spatial coordinates is used to represent a group of customers making up a block. Using a single set of spatial coordinates to represent a group of customers is better than using a set of coordinates to represent each individual customer. Thus, based on the team’s position within the network area, the application utilises the defining block’s spatial coordinates to display to the team the current context surroundings, including the network layout, and the relevant services required at that time. Information self-triggering user interface: Treating the network as geographically distributed blocks with defined spatial coordinates allows information pertaining to the blocks to be displayed as the vicinity of the block is entered. The system achieves this by matching the spatial coordinates of the user with the spatial coordinates of the block. This is termed triggering information by context [Brown , 1998]. The objective is to display the necessary information that the teams would require when specific tasks have to be performed. For example, upon receiving details of customers reporting faults from the Workforce Management Centre and accepting the task, the maintenance team enters these customer identifiers (e.g. a work-order number assigned to each customer fault) on the mobile device. These customers can be situated in the same block or lie scattered amongst different blocks. The application associates each customer identifier with the spatial coordinates of the block to which it belongs. The Business case for mobile computers Mobile application technology enables significant structural change in how organisations perform their tasks and accomplish business goals by moving existed business processes, whether automated or manual, beyond the organisation’s office to where-ever and whenever those can be tasked are carried out efficiently. In the recent past, one of the key obstacles to the successful deployment of mobile applications was the high cost to coordinate these tasks out in the field. Practice is already demonstrating that this obstacle has been overcome. The pay-off for mobile computing comes about through lowering the Total Cost of Ownership (TCO) of the technologies deployed and lifting the Total Benefit of Ownership of the mobile application. Lower Total cost of Ownership: Five years ago it was estimated by Intel Corporation [Intel, 2002] that the total cost of ownership difference between a notebook computer and a desktop was approximately $4000. By the year 2000 it had declined to less than $1000 and so the question was posed as to how much extra value does one need from a mobile computer user to make up this $1000 TCO differential. Depending upon the total cost of employment of the user itself, Intel estimated that only one extra hour of production per week to justify that TCO difference. With PDAs it could be the same, or less, so lowering the total cost of ownership is definitely feasible in this application. The reader can easily calculate the TCO applicable to its mobile application under consideration. In “Tapping the Power of the Mobile Enterprise”, a white paper from Extended Systems, some tips for determining the TCO are provided, covering direct costs of the necessary software licences and their maintenance, hardware costs, data security and access measures, and staff training and device deployment costs. Higher Total Benefit of Ownership: A key assumption is that mobile users are more productive. When Intel management investigated its mobile computing initiative, it had learned that mobile users realise productivity improvements of between three and eight hours per week. This analysis is borne out in most utility industries as well. According to a study conducted by the Gartner Group, business users with a mobile computer, and who spend 20 % of their time out of the office, realise a minimum annual benefit of 20 to 25% of their annual cost of employment through increased productivity and efficiency. Other benefits of this deployment, but incorporating PDAs and wireless access are quantitative (e.g. cost savings) and provide better responsiveness, better accuracy and better timeliness of, for example, network asset information. Also by having access to the corporate databases from field remote locations, the mobile user is assured of accurate and reliable information as stored in the Corporate repository. This ensures that the mobile worker can be certain that he/she has the correct tools and information (geospatial and other) for the task at hand. Conclusions The pay-offs for mobile computing applications are clear and the University has embarked on many different research projects, particularly involving the mobile workforce for utilities. It is anticipated that this work will assist African governments too in their dream of NEPAD and the task of bridging the infrastructure gap on the continent of Africa. References Baus J, Ding Y, Kray C, Walther U, 2001 "Towards adaptive location-aware mobile assistants," Workshop notes of the IJCAI 2001 Workshop on Artificial Intelligence in Mobile Systems. Brown PJ, 1998, "Triggering information by context", Personal Technologies, Vol. 2, No.1, pp.1-19. Cheverst K, Davies N, Friday A, 1998, "Developing Interfaces For Collaborative Mobile Systems," Proceedings of the First Workshop on Human Computer Interaction with Mobile Devices, University of Glasgow U.K, 21 -23 May 1998, pp.73-88. Cheverst K, 2000, "Some Strategies for Resource Sensitive HCI," Proceedings of Workshop on Resource Sensitive HCI, HUC 2000, Bristol. Cobb J, 2001, “Enterprise Mobile Solutions”, Wireless Development Services, SPRINT White papers, No 5 Davies N, Mitchel K, Cheverst K, Gordon B, 1998 "Experiences of Developing and Deploying a Context Aware Tourist Guide: The Guide Project," Proceedings of the First Workshop on Human Computer Interaction with Mobile Devices, University of Glasgow U.K, 21 -23 May 1998, pp.73-88. Intel, 2002, “Mobile PCs and Wireless: Business Users make the Productivity connection” Intel white paper. Intel, 2002, “Building the Foundation for Anytime, Anywhere Computing” Intel white paper. Johnson SD, 1998, “Productivity, the Workforce, and Technology Education”, Dept of Vocational and Technical Education, Univ of Illinois. Lancaster J, Alston D, 1996, “The inside story of outside PlantSpatial Data warehousing” AM/FM Australia-New Zealand AGM. NEPAD, 2002, “NEPAD@Work: Summary of NEPAD Action Plans” Rodden T, Chevrest K, Davies N, 1998, "Exploiting Context in HCI Design for Mobile Systems," Proceedings of the First Workshop on Human Computer Interaction with Mobile Devices, Glasgow, 21-23 May 1998, pp. 12-17. Schmidt A, Beigl M, Gellersen HW, 1999, "There is more to Context than Location," Computers and Graphics Journal, Vol. 23, No.6, pp. 893 –902. | ||
|
|